Ok, so lets elaborate on this idea a little, and acknowledge a few things about the current point system and associated medals.
Currently, acquiring points is the only criteria a person needs to be recognized via Medals on the forums.
While answers get you there faster, it’s not a criteria that a user have a single answer credited to them to obtain a medal. The current system benefits long term members, and encourages new comers to cheat and scam the system
to obtain points, which gets them medals. That system is clearly flawed. You can talk all you want about its suppose to encourage this, and do that, but mostly those arguements don't hold up given peoples dislike of the point system/medals overall.
Newbies/Rookies in baseball are measured on the same scale as the veterans (long standing members), by their stats. Their is no favortisms
to veterans on the stats. Below are a few common issues that people report and how stats could help solve those.
Issue: Un-moderated forums
Posts to un-moderated forums are not counted against your primary statistics for calculated against answers.
The issue here is whether we should allow/disallow answers from un-moderated forums to be included in primary statistics for answers.
Basically, I would say we calculate out all three combinations, at least until we identify which is the best indicator of a user that exemplifies some trait the forum values.
Points per Moderated Question (PPMQ) = Moderated Answers / Posts to Answered Moderated Question
Points per Un-moderated Question (PPUQ) = Un-moderated answers / Posts to Answered Un-moderated questions
Points per Question (PPQ) = All Answers / posts to all answered questions
Issue: Questions marked as answered but really aren’t.
In baseball terms this is basically a Reach On Error, and should be treated as such.
How? Simple, my suggestions would be to use one of the following
1)
Allow the moderator the option to mark the question as answered, but flag it somehow to indicate it’s not really answered (tentatively answered?).
2)
Implement an abandoned thread option to allow a thread to be quasi closed without an answer.
Statistically speaking this becomes a non issue.
Because the most meaningful stats will be based on an average, minor errors like this will only make up a small and progressively smaller percentage of users total points.
For this to be of any significance the moderator and users of the forums would not be doing their job.
In my opinion the primary reason moderators marks answers that are clearly not, is because they need to, to do their job effectively.
They have no choice, their job practically requires it.
So either acknowledge that fact, or give them other alternatives.
Issue: Questions left answered / abandoned
This is another Reach on Error issue, like the previous issue, I would implement an abandoned thread option to allows a thread to be quasi closed without an answer….if
this is really needed at all…
In my baseball stats solution this becomes a non-issue because posts to these questions are not counted against you, it’s a free 2 points, but it has no effect on
any of your stats. This is because ultimately posts to a moderated question don’t count until the question is marked as answered.
This however would modify my previous at bat definition as follows.
At bat = Post to an
answered question in
a moderated forum. Until the question is marked as answered, posts to them are not counted when calculating an at bat.
You will also notice that I made this distinction on my PPMQ/PPUQ/PPQ definition.
Issue: Disscussion’s don’t have answers
All posts to discussions are a free 2 points, but don’t count against stats based on Answers.
You could have stats based purely on discussions and Vote Up points, but unless they prove to be of significant value, i think they would be of little value.
Issue: Moderator errors and marking incorrect or to many answers
This is basically the same as the
Questions marked as answered but really aren’t issue.
Again, statistically speaking this becomes a non issue.
Because the most meaningful stats will be based on an average, minor errors like this will only make up a small and progressively smaller percentage of a user’s total points.
For this to be of any significance the moderator and users of the forums would not be doing their job.
Issue: Troubleshooter vs Cherry picker
Someone that spends a lot of time on difficult problems requiring a lot of posts will automatically have a lower Batting Average (Answering Average) and associated PPMQ/PPUQ/PPQ.
While someone that Cherry Picks ease questions would have a higher Answering Average.
First, as many users have pointed out in posts wanting to get rid of the point system totally, this type of concern shouldn’t be your primary motivation to be using
the forums. If it is, you should move on.
Secondly, what we are talking about are Roles people take on, usually based on their skill.
These roles are neither good or bad, or preferential over another.
If you want to troubleshoot, then do that, if you want to cherry pick, do that, if you want to do both, than do that.
You can’t get rid of this behavior, you can only choose to reward or discourage it.
Personally I would consider rewarding them both, thus showing no preference.
My suggestions to reward both types of users
1)
Identify and Reward the
troubleshooter
a.
This is done by classifying a question as troubleshooting question. The simplistic way would be based on the number of responses either in total (all users) or by a single user.
Example: more than 15 posts by all users, and/or more than 5 posts for an individual user.
b.
Once you have a method of identifying troubleshooting questions, you can create a whole set of statistics based on that classification.
TroubleShooting Answering Average = Answers to troubleshooting questions / Posts to answers to troubleshooting questions
2)
Identify and Reward the
cherry picker
a.
Obviously this is inherent in the design of statistics, given that all stats will reward the quick answer.
b.
For example if you identify cherry questions, as ones answered in a single post by any one user and no more than 3 posts total.
Cherry Answering Average = Answers to cherry questions / Posts to answers to cherry questions
c.
Although, I think this would be an interesting stat I am not sure of it meaning fullness, for all we know people that engage in this activity may not be very good at it anyway.
